Emerald, the plaintiff in this diversity suit (governed by Illinois law) for breach of contract, obtained a verdict and judgment for $1.1 million against the defendant, Allmerica. Allmerica contends that Emerald should not have been awarded any damages apart from the cost of a $150,000 surrender fee discussed later in this opinion. Emerald, cross-appealing, wants greater damages than the jury awarded; but on the view we take of the case, the cross-appeal is academic.
Allmerica sells variable annuities both directly to annuitants and to intermediaries who resell to annuitants. An annuity is in effect a reverse life-insurance contract: you pay a lump sum to the insurance company in exchange for a promise to pay you an income for life; the longer you live, and also the higher the return from investing the lump-sum purchase price (that in
Emerald, which one might have thought an intermediary customer, in March 1999 bought $5 million worth of variable annuities from Allmeriea and later increased its investment to hundreds of millions of dollars. Emerald was not interested in reselling its variable annuities to prospective retirees, however. It wanted to engage in arbitrage. An arbitrage opportunity arises when the same thing is being sold at two different prices and the difference is due to some oversight or other error, or to price discrimination (charging different prices for the same good or service on the basis of different intensities of consumer demand for it), rather than to costs of transportation or other circumstances that might place the good in different markets and thus prevent uniform pricing. The arbitrageur spots the artificial price difference, buys at the lower price, and resells at the higher price. The effect is to bring about price uniformity, which terminates the arbitrage opportunity. Arbitrage is a socially useful activity because if the same good or service, costing the same and traded or tradable in the same market, is selling at different prices, one of those prices is too high (excluding the case in which one of the goods is selling below cost, in which event the price is too low) from the standpoint of an efficient allocation of resources.
Emerald had noticed that identical securities were selling at different prices in mutual fund accounts offered to the purchasers of Allmerica’s variable annuities. The mutual funds set the prices that they charge for shares in their funds at 4 p.m. New York time, which is when the New York Stock Exchange closes. Those prices are a composite of the prices of the shares (in publicly held companies) owned by the fund. (On the pricing of mutual fund shares, see generally SEC, “Disclosure Regarding Market Timing and Selective Disclosure of Portfolio Holdings: Proposed Rule,” 68 Fed.Reg. 70,401 (Dec. 17, 2003);
DH2, Inc. v. SEC,
To take advantage of the discrepancy between the composite price and the prices of the fund’s constituent assets, Emerald would buy shares in the mutual funds minutes before it was 4 p.m. in New York (so as not to attract imitators, who would bid up the price of the shares in their eagerness to buy them) and sell them the next day, or within a few days, once the price of the foreign-traded shares was reflected in the price of the mutual fund shares. See also
Kircher v. Putnam Funds Trust,
Emerald financed its purchases of shares in the Scudder International Fund by transferring money from the Allmerica money-market fund in which it parked its investment in annuity contracts when it was not engaged in arbitrage. When the contract with Emerald had been made, Allmerica had allowed buyers of its annuities to transfer their investments to any other mutual fund with which Allmerica had an arrangement. But Emerald’s frequent transfers between the money-market fund and the Scudder International Fund were a pain to both Allmerica and Scudder. The transfers were large — as much as $111 million. Scudder had to keep a large amount of cash on hand, which it would have preferred to invest, in order to redeem shares in its fund when Emerald, having bought the shares because it believed them underpriced, decided soon afterward to return to its money-market fund. Scudder’s other investors suffered and so therefore did Allmerica, since its variable-annuity contracts lost value.
Had Allmerica known that Emerald was buying variable annuities in order to engage in international time-zone arbitrage, it would not have sold to Emerald, at least in the quantity it did; other sellers of variable annuities had stopped dealing with Emerald. In December 2001, Allm-erica limited the number of transfers that its customers could make from the Scud-der International Fund to another account to one per month. That action precipitated this suit. To add insult to injury, when Emerald later withdrew its money from Allmerica, Allmerica charged a $150,000 surrender fee, which Emerald had to pay to get its money out. The district judge ruled that the imposition of the limit was a breach of contract, and Allmerica does not contest the ruling.
In July 2004, after this suit was brought, Allmerica, perhaps anticipating the district judge’s ruling on the transfer limitation, closed the Scudder International Fund (and other international funds in which market timing was likely to occur) to new investments. With the international funds closed, no longer could Emerald transfer money into them from the money-market fund. The district judge ruled that this method of stopping market timing was not a breach of Allmerica’s contract with Emerald — which makes one wonder whether any damages should have been awarded for the acknowledged breach of contract noted in the preceding paragraph. A breach of contract to be actionable has to cause the plaintiffs injury. Had Allmerica not broken its contract, it almost certainly would have done what it did when it was sued for breach — closed the fund to new investments — so that Emerald’s loss of profits from arbitraging would have been the same.
Allmerica doesn’t make that simple argument, however. Instead it argues that the damages awarded by the jury were unforeseeable, and alternatively that they were hopelessly speculative. The first argument relies on the venerable precedent of
Hadley v. Baxendale,
9 Ex. 341, 156 Eng. Rep. 145 (1854), which, as we noted in
Evra Corp. v. Swiss Bank Corp.,
Allmeriea overreads Hadley and the cases following it. They are cases about special handling. The Hadleys owned a flour mill. The millshaft broke, and the Hadleys hired Baxendale to transport the broken millshaft to a shop that, using the broken millshaft as a model, would make a new one. Because the Hadleys had no spare, the mill was shut down until the new millshaft arrived, and they incurred substantial losses. The receipt of the new shaft was delayed as a result of a breach by Baxendale of its contract of carriage. The court held that the Hadleys could not recover the profits they had lost because of the delay. Had they wanted Baxendale to take special care to get the new mill-shaft to them by the contractually specified deadline, they should have negotiated for that care, that special handling; undoubtedly Baxendale would have demanded a higher price.
An Illinois case illustrates this point. The plaintiff in
Siegel v. Western Union Telegraph Co.,
This case isn’t like Hadley or Siegel or our hypothetical photographer’s case. It is not that Allmeriea failed to take special care to fulfill its contractual obligations because Emerald had failed to negotiate for special care. This is a routine case in which the victim of a breach of contract is suing for the profits that he thinks he would have made had the breach not occurred. A buyer is not required to disclose the profit he anticipates from dealing with the seller; such a requirement would kill the incentive to seek out profit opportunities. There would be no incentive to engage in arbitrage if the arbitrageur had to disclose to the person from whom he was buying an underpriced good that it was underpriced because it could be resold in another market at a higher price. For then the would-be seller would resell it in that market himself, bypassing the arbitrageur and pocketing the arbitrageur’s anticipated profit.
The old but still good case that governs here is
Laidlaw v. Organ,
In support of its second ground for attacking the judgment — that the damages awarded, though modest in relation to what Emerald had sought, were speculative, or in other words not proved — Allmerica finally reaches solid ground. Emerald presented just two types of damages evidence. One, excluded by the district judge, consisted of two reports by a finance professor, Steven Buser. The other consisted of testimony by a principal of Emerald about trades that he would have made had Allmerica not pulled the plug in December 2001.
Buser’s initial report proposed that if permitted by Allmerica to continue its market-timing trading, Emerald would have earned an annual rate of return on its investment of 34 percent for 20 years, for a discounted present value of $150 million. That was a preposterous estimate, properly excluded by the district judge under Fed.R.Evid. 702. The essence of an arbitrage opportunity is that it is transient. It is the exploitation of an economic anomaly, and the process of exploiting the anomaly eliminates it (that is the social function of arbitrage). If a pharmaceutical company sells drugs at a lower price to hospitals than to drugstores, then the hospitals, unless restricted, will order more drugs and sell the excess to drugstores at a price intermediate between what they pay the pharmaceutical company and what the drugstores pay. As the company’s revenues from selling to drugstores decline, it will have to reduce its price to them until it is charging them the same prices it charges the hospitals; otherwise it would have no more sales to drugstores. So eventually there will be just one price. Likewise in this case: by the time of trial the process of eliminating the arbitrage opportunity that Emerald had exploited was well under way. This was due in part to a rule proposed by the SEC, “Disclosure Regarding Market Timing and Selective Disclosure of Portfolio Holdings,” supra, that flagged the type of arbitrage Emerald was engaged in, and that when adopted (as it was on April 23, 2004, 69 Fed.Reg. 22,300) eliminated the arbitrage opportunity by requiring funds to use a different method of pricing their shares; and in part to the refusal of more and more mutual funds to do business with market timers. See generally Investment Company Institute, “Questions and Answers About the Mutual Fund Investigations” (Nov.2003), www.ici.org/issues/ timing/arc-reg/faqs_timing.html (visited Jan. 30, 2008). Buser’s choice of a 34 percent annual rate of return was a blind extrapolation from Emerald’s history. It ignored the changed circumstances that we have mentioned, even though they had already caused the company’s rate of return from trading in the annuities it had bought from Allmerica to decline from 40 percent to 14 percent in the year of the breach of contract (2001); the rate of return actually turned negative before trial. And the 40 percent rate had doubtless reflected not just Emerald’s trading strategy but also the stock market boom of the 1990s, which ended in 2000.
Buser’s first report was so irresponsible as to justify the judge’s decision to exclude the second report summarily. Buser had demonstrated a willingness to abandon the norms of his profession in the
This leaves the testimony of the Emerald principal. With the benefit of hindsight, he picked days on which he said he would have traded during the damages period (which the judge had cut off at July 2004) had it not been for the breach of contract, which prevented him from doing so. His testimony was as worthless as Siegel’s would have been had Siegel picked the horse to bet on after the race had been run. Once the price of shares, traded on a foreign stock exchange, at the opening of the New York Stock Exchange is known, one knows with certainty whether buying shares in a mutual fund that owned those foreign-traded shares at 3:59 p.m. New York time the previous day would have been profitable. That is no evidence that one would have traded then, for there is no way in which such testimony could be tested. It would be different had Emerald based its market-timing trading on a formula that determined when and in what dollar amount to transfer money into the Scudder International Fund and when to transfer back out of it into the money-market fund. But it did not use a formula. It traded on hunch.
Emerald did present some evidence of trades it made during the damages period involving international mutual funds not offered by Allmerica, but whether it would have upped the ante by trading in the Scudder International Fund as well was unproven. Before it was frozen out, Emerald often had traded in other funds without also trading on the same day in the Scudder fund.
What we have said is enough to require that the award of damages be reversed. But for completeness we respond to two further arguments pressed by Emerald. The first is that Allmerica, having broken the contract with it in December 2001, could not later invoke the contractual provision entitling it to close its accounts to new money, which truncated its liability at July 2004. In defense, Allmerica invokes the doctrine of “partial breach.” Like many legal doctrines, it is badly named. There is no such thing as a partial breach. There is a breach of contract, and there are alterations and terminations that are not breaches. The doctrine is really an aspect of election of remedies. If a party to a contract breaks it, the other party can abandon the contract (unless the breach is very minor,
Circle Security Agency, Inc. v. Ross,
The judgment is affirmed with respect to liability but reversed with respect to damages. The district court is directed to enter a judgment that Allmeriea broke its contract with Emerald in December 2001 but (since nominal damages are not sought) that Emerald is entitled to no damages or other relief beyond the $150,000 surrender fee, plus appropriate interest.